What is grid computing ? “Grid” is a commonly associated word that is lumped in with scientific and techno babble, and used heavily in science fiction to make things sound technical or advanced. So, people automatically expect grid computing to be something complex and probably abstract. They’re in for a shocker, then, because while there’s a few definitions for this term, all but one are just alternate definitions for concepts we’re already familiar with.
So, what is cloud computing all about? Well, like I said, there are a few definitions for the concept, so we’ll touch on a few of them. Again, all but one, you will find, are things you’re familiar with by other names, and where I remember to, I’ll cite a likely alternate term you know for each definition.
The first familiar definition for grid computing is any form of remote computing wherein a series of servers, linked together in a logical manner at a single location, form a super computing body for the cloud deployment model. This is actually the oldest form of arrayed computing, and the combined and synchronized task-oriented units add up to fierce parallel computing power.
The oldest term for this is a server farm, but the newest term is a grid topology cloud architecture.
The second meaning of grid computing is in the form of how memory and threads are managed in high-powered multi-core machines to achieve results PCs aren’t normally rated to do. It’s a similar concept to a server farm, where each thread -a series of tasks being performed, are managed in a logical matrix which serves as a sum of parts identity.
But, the term you’re hearing the most for grid computing is a bit more complex, but not terribly so. The buzz word form of grid computing pertains to a management system where a series of computing resources from disparate but coordinated domains work together to achieve a common goal.
What this changes is, it eliminates the need for a data center, but it’s still more structured than cluster computing. Various domains from across the world can be tied together to handle specific aspects of a computing task, and have alternate locations as redundancies to account for outages and distances between units and users.
With domains being able to entangle this way, organizing what servers do what, which would be a challenge without a data center otherwise, becomes much easier. When something is easy to maintain, that means iti s easier to keep it working.
For the moment, even in the future nothing works, so if they can get rid of the rigidity of data centers, while preventing the anarchy of cluster computing, then I for one am all aboard.
This kind of ingenuity will only serve to further cement SaaS and cloud computing as a viable and very real deployment model for the future. The world is a global place, so our computing should be a more durable global entity as well. It looks like most of the scientists are on the same page as I am here, as well.
So, what is grid computing? I’d say it’s the future.